Should cars intervene on our behalf to keep us safe?

Malcolm R makes a good point about automatic transmission being a safety feature.
I also have manual paddles fitted which allow manual override of the automatic selection (on mini cooper).. I would not want to be without these in a future vehicle. I find them really useful.

What you’re partially talking about is adaptive cruise control, Alfa. Nice to use on motorways and especially good for avoiding speed tickets when in built up areas and you’re having to watch both the multiple hazards and the speedo.

One of our Toyotas is fitted with a lot of devices – but the Lidar display which houses many of the sensors sometimes goes into shut down mode on cold days 🙂

Most accidents don’t occur through driver boredom but through driver error, and last year there were 140,086 road traffic accidents in the UK that resulted in personal injury. That works out at almost 12,000 a month, or just under 400 every day. The top five causes:

1. Driving under the influence of alcohol and drugs
2. Tiredness
3. Excessive speed
4. Distraction – mobile ‘phones, changing radio channels.
5. Failure to react to severe weather

Interestingly, fully autonomous cars could cope rather well with all those issues.

Do autonomous cars automatically give you a breath test?

And how do they cope with fog, snow, black ice?

My car is 19 years old so rather lacking in recent modern technology, but it has managed to stay out of trouble so far with me in full control.

Interestingly, there are devices that can be part of your car that do provide breath testing, but the weather is the real reason these developments could be invaluable, since fog won’t provide an impediment to the sensor arrays. Snow could stop them – as it can with any ordinary car, and which is why we have a Toyota 4 x 4 as well, which goes up any snow-clad incline with relative ease – and the black ice question is another reason, really, why the fully autonomous vehicle should be a lot safer, since its in-built accelerometers will detect any slight drifting well before a human driver could react, and correct accordingly.

I’ve always been a technophile, so it’s worth viewing my enthusiasm in that light, and Duncan does make some very sage points about location tracking and even hacking, as cars have been unlocked (but not, as far as I know, actually started) remotely across continental distances.

But the real reasons I await their arrival with great enthusiasm are the freedom they will afford to the disabled in remote areas and the increased safety they should provide for everyone on the roads.

The theory was put forward by a New Scientist columnist who argued that because the full autonomous car would have to be – for all intents and purposes – almost entirely foolproof in safety terms, once people began to be able to identify them as such they would come to realise that the normal rules don’t apply and would simply walk across roads as and when they pleased.

I think it has some merit, in fact. The irony would be that because the fully autonomous vehicle was far safer than the human-controlled, that characteristic would be exploited by the opportunistic.

A bit like teaching your children to drive! We like to think of technology as “foolproof” but it is not. I’d prefer it if we put a lot more effort into public transport rather than individual vehicles that demand more road and parking space. And research into better and lighter electric storage for all vehicles.

I’d like to see a future where the major road users – commuters – no longer have to spend time travelling, wasting time, to places of work by putting work nearer where they live, and using local offices instead of a single headquarters. Get unnecessary traffic off the roads.

While I agree (and wonder, frankly, why more hasn’t been done) about public transport in urban areas, in many rural and mountainous areas in some ways it’s a non-starter. Don’t get me wrong; it exists, but travelling by horse and cart might be slightly speedier.

However, if (as many companies are suggesting) London becomes something of a commercial wasteland following Brexit then the problem might solve itself. Of course, we’ll then have to be caring for the millions unemployed, but I suppose we can’t have everything.

It’s an intractable problem: commerce tends to flock together and everything else follows.

John: yes – that’s the first time I’ve read a seriously concerning issue surrounding autonomous vehicles. What I can see happening is that the first place autonomous systems will be deployed will be in the stopping arena, in part for the very reason you specify. I would imagine that the creators and designers of AVSs (autonomous vehicle Systems) will already have considered this and will have realised that a shunt is a real possibility, so legislation would be needed to amend the Construction and Use regs to enforce mandatory stopping communication systems between all vehicles, and not simply between the AVSs fitted. It’s really the only way the system can work safely.

The one thing at which the UK excels is technical innovation and software. The Government seems very keen to push the areas concerned, so will, I suspect, be only too willing to legislate accordingly.

Of course, we can also dream that one day inertialess braking will become a reality, in which case we can stop dead when necessary.

There are around 26 million vehicles on UK roads, and their average age is 7.8 years. It will be impossible to fit all these vehicles with devices compatible with automated vehicles. We will have to deal with mixed traffic for many years to come, given the increased longevity of vehicles. I’d expect if we do introduce autonomous vehicles they would be restricted to particular areas and probably segregated from “normal” traffic.

That’s also a possibility, but simple figures can be misleading. There are around 31m cars on the road but there’s also around a third of a million HGVs which, together,. move around 75% of goods around the UK, when rail should certainly be moving most. That’s an aside, however, and while I agree that it’s unlikely that retrospective legislation would be enacted for 100% of vehicles it can certainly be enacted for most and auto braking can be retrofitted comparatively cheaply. John was talking about the instance of the HGV specifically, and I would imagine that legislation pertaining to them will be enacted.

Separation by a highways version of air traffic control using GPS is a possibility. LGV’s and delivery vans are increasingly in communication with their operations bases. By 2025 onwards significant progress could be made towards autonomous driving on major motorways. On exit a transponder would switch off the autonomous control and the human driver would regain control. In urban areas one advantage of autonomous driving would be speed control: no more excessive speeds in residential roads or late-night boy racers tearing up the high street. The emergency services would have to have an opt-out.

And think about this – a computer-managed school run with all movements safely coordinated and synchronised! No more late-for-violin lesson anxiety!

Now there’s an offbeat idea…

And no more speed bumps? 🙂

There would be no need for speed bumps, chicanes, and speed cameras. No parking contraventions on the highway. Disabled bays always clear of obstruction. Roundabouts replaced by signals for pedestrians and inter-linked with guided traffic flows. Buses and taxis given priority at junctions and programmed to make turns banned for other traffic. Electric cars could even book their recharge at a charging station en route. When setting off and loading the destination coordinates and preferred arrival time the system would check the whole journey for congestion and delays and plan the route not just by geography as now but by using all relevant data. Platooning would be directed for optimum efficiency and economy. The module in the car would become a slave to the overall system. Fancy, being in the right place, in the right lane, at the right time, at all times; motoring doesn’t get much better than that. It would be satnav in excelsis.

Dream on John. 😴🚗😴🚍😴🚘😴🚋😴🚌😴

Utopia of the future maybe, but we are unlikely to experience it in our lifetimes.

You would not even need anyone in the autonomous car. Just send it on its way to collect the groceries, deliver your report to the office, take granny out for spin. We could just sit on the sofa at home watching the tv channels automatically selected for us by the Data Collection Authority.

In wonder who would be held accountable for accidents when the foolproof technology is fooled?

For this week’s work expedition “up north”, our hire car was a nice comfy Skoda Octavia.

This came with the full “no frills package”, i.e. no satnav, no cruise control, no parking sensors and only a manual handbrake.

I’m really beginning to like those Skodas…

Lovely car, the Octavia. Wish we’d bought one instead of an Audi. One of the few cars with plenty of legroom and comfort in the rear seats plus a vast boot. Our friends’ Jaguar, for all its length, is surprisingly cramped.

Not as good as an Alfa 🙂🙃🙂🙃🙂

…red with matching rust? 😉

Not red and NO rust. That was an optional extra over 20 years ago.

100% agree with you wavechange except I don’t like cruise control and don’t have it on my car anyway.

I find mirrors an essential safety feature.

Mirrors are simple but effective, but the manufacturers want to introduce blind-spot warning systems. Many of us cope well without them. Some cars seem better than others regarding blind-spots so maybe there is a case for copying good design.

I rarely use cruise control as the motorways and other major roads we generally frequent are not good candidates. However, I support blind spot help – overtaking vehicles can surprise you sometimes when they are not quite past; I’ve seen a number of near misses when the vehicle about to be overtaken has started an overtake manoeuvre itself , abruptly abandoned.

It’s quite a hot topic in most places, Duncan. It’ll be interesting to see how it all turns out.

Given that it was the owner’s choice to operate an autonomous vehicle I would expect liability for an incident or collision to attach to the driver. The lawyers will have a field day. Perhaps the government, in its enthusiasm to get this show on the road, will change the law and make the driver liable every time without dispute. But that would put many people off owning an autonomous car.

The driver of an autonomous vehicle will not be able to take their eyes off the road and must continue to look out for hazards, and they must be ready to take back control instantly. The theory of automated separation is wonderful until a deer comes crashing across the road – or will they also have to have a plug-in module to broadcast advance warning of their movements?

I don’t believe the Deer scenario is even mildly problematic for the designers, John; animal collisions (random interceptions) was almost the first item on the snaglist.

I believe the truly autonomous vehicle will lead to a huge improvement in road safety, frankly. Having (in a voluntary capacity) worked with the IAM locally and helping assess drivers who were wondering about taking the test, as far as I’m concerned the sooner car control is removed from drivers the better.

It’s easy to believe we’re all better at driving than an automated system, and easy to believe automated systems haven’t yet approached the level of human ability. But honestly the human ability I see, day in, day out makes me wonder if some of the drivers ever attended a driving lesson.

We’re busy finding all the faults at the moment but consider this: the automated vehicle will never take chances, never become stressed while taking the children to school, never become upset at the behaviour of other road users, never take a corner too wide nor a bend too quickly, never park badly, never become bad tempered, irritable, tired, sneeze, cough, blink, get distracted by a wasp of become afflicted by any of the thousand things which flesh is heir to. ‘Tis an automation devoutly to be wished.

If the driver has to do all the things you say John then I dont see this being popular nor would anybody trust a motorised Digital Chip when GPS tracking can already be intercepted and interfered with . Also the control mechanism would need 100 % shielding from outside transmissions , the FBI and other have already proved interception can take place as well as all the US police forces that use RF /GPS tracking control of Bait Cars . There is no way on earth I would trust it in its present state its been hacked and can be hacked and all your data of where you go etc will be used to the profit of third parties -a disaster waiting to happen( for the owner ) if your version of traffic laws apply .

A few years ago a deer came out from trees that bordered a country road and collided with the front of my car. A car would not have stopped in time whether “autonomous” or not. And it could equally have been a child on a bike. Decent drivers will be aware of potential hazards like this and drive cautiously. I think there is faith being place in technology that is not justified (well, not yet). I also wonder whether many people want it. Perhaps if we could make trains totally automatic first? There are more constraints on them – rails, segregated track, automatic signalling, remote tracking – and yet we still rely on alert drivers.

Generally I agree with you, Ian. I think safety will be greatly enhanced, but not before there is very widespread uptake of the technology and it might come in the commercial vehicle sector first.

I don’t share Duncan’s concerns about unauthorised data access – I can’t see why anyone would want to do it and I wouldn’t feel personally at risk. I cannot believe there will not be sophisticated encryption and complete data isolation between the vehicle registration mark [number plate] and the on-board controller. I don’t care who knows where I’ve been because I’ve got nothing to hide – or it’s just another alibi if needed.

I agree with Ian’s comments on driver competence. I was recently waiting for a friend at a traffic-signalled road intersection. I was on one corner of the junction and, with some time on my hands, was observing the traffic. A number of vehicles were turning left off the major road alongside me to enter a side road of restricted width and with vehicles queuing to emerge when the lights changed. I could not believe the number of vehicles that clipped the corner, some by as much as a metre, because they were in the wrong gear for the manoeuvre and going too fast. A number of the vehicles were mini-buses taking children to school or carrying elderly or disabled passengers. Some vehicles swerved so far over into the side road that they had to brake fast and make a rapid steering correction to avoid a collision with the first vehicle in the queue – yet that was probably a route they took regularly.

Ian – I should be interested to know the prevailing view on traffic congestion. If every vehicle proceeds at the correct speed with the correct separation my intuitive response is that our roads cannot cope: it is only bunching and hot-foot accelerations from time to time that keep things flowing. The idealised world of autonomous driving could produce continuously moving gridlock in heavy-traffic areas and at peak times. On the other hand, the total discipline of autonomous control could actually free up road capacity and enable more vehicles to travel along the same sections of road, and with more advanced traffic signal control responsive to route demands and the volumes of vehicles desiring each approach lane the days of the blocked roundabout and other choke points could be over. It could reduce or postpone the need for expensive new roads and capacity-enhancing modification schemes that themselves add to congestion while under construction.

I don’t understand why you so much faith in digital electronics John . All the advertising saying digital banking “safe ” – digital internet transactions “safe ” – digital this or that “safe ” have proved top be wrong . Here is a quote from the Washington Post – Hacker disables SUV on busy highway. Andy Greenberg was travelling along a busy Interstate in St.Louis when he suddenly lost control of his vehicle -the accelerator stopped working -Greenberg began to panic it stalled . No he wasn’t hit by another car , nor engine trouble – it had been HACKED .An aside article – auto makers rush to add wireless features , leaving our cars open to hackers exactly what I was talking about . There is no “pure ” transmission of data unless you are talking about military communications and even there the US (at least ) has armed satellites to zap other foreign satellites as well as laser interference its a big topic in military circles . Trust in robotics ? not at this moment John. I have many more of the same. Just look at “digital dropout ” in Wi-Fi in the home and that’s at close range Big Arnie has a lot to answer for in his robotic movies . Can you find a website that guarantees no accidents caused by automated vehicles ?

It is not just the electronics that bother me, duncan, but our ability (or lack of) to design instrumentation and software to consider all conceivable possibilities without error. I am also unsure as to why universal autonomous vehicles are thought necessary. The consequences of equipment failure exist, just as do failures in conventional vehicles and their drivers. I am all for giving drivers additional aids to help the driving workload, but still prefer to give ultimate control to a person, not a complex electronic system where we rely totally on the standards being properly implemented by all the manufacturers who will have to become involved – worldwide.

In all the years I have driven, I have never witnessed a serious road accident happening. Despite traffic increasing, fatalities have halved since 2005, and from what I read now stand, for cars, at around 1.8 per billion vehicle miles. The most vulnerable groups are cyclists – pedal 30.9 and motor 122.3. I don’t know how these forms of transport will be made autonomous to avoid accidents, other than through those involved with cars.

However, clearly any measures that can sensibly be adopted to reduce non-fatal and fatal accidents, will be encouraged. But are autonomous vehicles then the best way? Or is their other motivation behind their promotion – is it technology for its own sake to profit commercially from its adoption. Perhaps the Government have some cost benefit analysis to explain their proposal?

Well said malcolm – I agree .

That is where I stand on it, Malcolm. I can see potential advantages in a fully automated and inter-communicating road transport scenario but am worried about the interim position. As I wrote previously, I think further advances in driver assistance features will make autonomous cars unnecessary. Given that a driver will have to remain at the wheel there is a concern about loss of concentration and sensory detachment from the conduct of the vehicle. Perhaps Ian has more information on this.

For me this discussion is actually academic because I can foresee no circumstances in which I shall be using an autonomous car. It would be good if some younger members of the community gave us their views on this as they will be the first to experience it in real life.

On your last post, Malcolm, nothing you say in it describes autonomous cars exclusively. Exactly the same things can happen to human drivers.

Where I see autonomous vehicles being used most is in the disabled community initially; in our nearest town convoys of mobility scooters merrily mow their ways through the assembled pedestrians, and they travel from the homes to do so. But with fully autonomous cars the disabled could be afforded a much fuller life, and might not have to leave the rural villages they often love but have to leave, because of a poor transport infrastructure. In particular, the development would be a major life changer for the visually disabled.

I think assisting the driver is good but taking over from the driver seems a step too far at the present state of development. Once all the potential ‘assistance’ features have been provided total automation seems unnecessary as it introduces additional risks.

Here is an example of what can go wrong: https://www.theguardian.com/technology/2016/sep/26/google-self-driving-car-in-broadside-collision-after-other-car-jumps-red-light-lexus-suv

I suppose that self-driving cars have the advantage that there will be no drivers to be distracted by making or receiving phone calls.

The car in the top picture looks like a Google StreetView camera car, or is the apparatus on the roof part of its autonomous observation and navigation system?

I notice a plethora of power lines alongside the road in the top picture. I hope the on-board electronics in autonomous cars will be fully immunised from any interference.

In the background of the second picture there is an “Interstate Batteries” van with a deformed hood being reverse-loaded onto a tow-truck. Judging by the impression made in the metalwork of the Google car the van was the likely cause of the impact which must have been fairly forceful – yet the car apparently remained upright, no glass appears to have been broken, the airbags deployed, and no injuries were sustained. This is testimony to the much improved safety characteristics of modern cars, and even the van has a ‘crumple zone’ to reduce the kinetic energy of the crash. In the circumstances described in the article it is unlikely that a human driver of the car would have been able to avoid a worse collision and might possibly have swerved risking impact with other vehicles or people.

It’s not really an example of what can go wrong. Red light jumpers are extremely dangerous to vehicles and I agree that a human driver would likely have fared no better. Of course, enhanced proximity detection and assessment of approach speeds might cause the AV to hold back, as it sensed the approach of a light-jumper.

Yes, that’s interesting. I can only imagine that the pursuit of this new technology registers more with Americans than with Europeans who appear to be generally indifferent.